To further our understanding of biological processes and disease mechanisms, it is important to consider gene functions in the context of complex macromolecular networks. For such "systems" approaches to be developed, physical protein-protein interaction, or "interactome", networks need to be systematically mapped, in addition to regulatory and metabolic networks. Similar to the Human Genome Project, the feasibility and usefulness of a "human interactome project" was first investigated using model organisms. Early versions of interactome maps are available for S. cerevisiae, C. elegans (see below) and D. melanogaster at proteome scale. The ultimate goal of the C. elegans interactome mapping project should be to determine all or nearly all physical protein-protein interactions that can occur in all cells, at all stages of development, and in standard environmental conditions. Here we propose to expand the worm project from its recently released version (W15), containing 5,500 potential interactions obtained by testing a matrix of -2,000 by -14,000 proteins and representing approximately 5-10% of the total interactome. Our goals are to improve completeness, coverage, and data quality of the worm interactome map. We estimate that in 3 years, we will be able to map -15,000 additional high confidence interactions, use an independent co-affinity purification (co-AP) protein interaction assay to systematically validate these Y2H interaction and finally, systematically compare gene expression patterns of gene pairs corresponding to protein interactions. Our specific aims are to: i) improve the completeness and coverage of the worm interactome map, ii) improve the quality of the worm intearctome map by globally addressing the issues of both technical and biological false positives, and iii) develop and apply computational tools to release, visualize and analyze the interaction data.